首页 > 论文 > 光学学报 > 38卷 > 6期(pp:610003--1)

基于卷积神经网络的立体图像舒适度客观评价

Objective Assessment of Stereoscopic Image Comfort Based on Convolutional Neural Network

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

基于卷积神经网络模型,提出一种立体图像舒适度评价方法。该方法无须提前根据特定的任务从图像中人工提取具体的特征,而是模拟人脑处理机制对图像进行层次化的抽象处理,自主提取特征。该方法采用三通道卷积神经网络结构,分别对原始图像进行主成分分析,以及32×32、256×256两种尺度的分块处理得到三条通道的输入数据集,根据输入数据设计每条通道的网络结构。采用两种尺寸分块处理得到不同尺寸的图像块特征信息,采用主成分分析降维处理得到原始图像的整体信息。此外,通过随机丢弃、局部响应归一化等方法提升算法的评价性能。实验结果表明,以修正线性单元为激活函数、输出层用Softmax分类器,对天津大学TJU立体图像数据库中400幅不同舒适度等级的立体图像样本进行测试,等级分类率正确达94.52%,优于极限学习机、支持向量机算法。

Abstract

We propose a new method for stereoscopic image comfort assessment based on convolutional neural network, which does not need to extract specific manual features from images in advance according to specific tasks, but simulates hierarchical abstract processing mechanism of human brain to extract image features autonomously. This method adopts three channel convolutional neural network structure, and the input data sets of the three channel are obtained by reducing the dimension of the original data samples through principal component analysis, and chopping the original data samples into two size image patches (32×32, 256×256), respectively. The network structure of each channel is designed according to the input data sets. In addition, the classification accuracy of this method is improved by introducing dropout and local response normalization, etc. With rectified linear unit as the activation function and Softmax as the classifier in the output layer, experiment results on 400 stereo image samples in TJU database with different comfortable levels show that, the correct classification rate of this method is 94.52%, which is higher than that of the extreme learning machine and support vector machine.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TN911.73

DOI:10.3788/aos201838.0610003

所属栏目:图像处理

基金项目:国家自然科学基金(61520106002,161471262)

收稿日期:2017-12-01

修改稿日期:2018-01-03

网络出版日期:--

作者单位    点击查看

李素梅:天津大学电气自动化与信息工程学院, 天津 300072

联系人作者:常永莉(cyl920611@163.com)

备注:李素梅(1975-),女,博士后,副教授,硕士生导师,主要从事立体信息处理和计算机视觉方面的研究。E-mail: tjnklsm@163.com

【1】Hou C P, Ma T T, Yue G H, et al. Multiply-distorted image quality assessment based on high-order phase congruence[J]. Laser & Optoelectronics Progress, 2017, 54(7): 071001.
侯春萍, 马彤彤, 岳广辉, 等. 基于高阶相位一致性的混合失真图像质量评价[J]. 激光与光电子学进展, 2017, 54(7): 071001.

【2】Yang J C, Hou C P, Shen L L, et al. Objective evaluation method for stereo image quality based on PSNR[J]. Journal of Tianjin University, 2008, 41(12): 1448-1452.
杨嘉琛, 侯春萍, 沈丽丽, 等. 基于PSNR立体图像质量客观评价方法[J]. 天津大学学报, 2008, 41(12): 1448-1452.

【3】Zhu Q S, Zhi L O, Liu R, et al. Research on image conversion from planar into stereo[J]. Computer Science, 2007, 34(7): 225-228.
朱庆生, 支丽欧, 刘然, 等. 平面图像立体化关键技术研究[J]. 计算机科学, 2007, 34(7): 225-228.

【4】Wang Z, Bovik A C, Sheikh H R, et al. Image quality assessment: from error visibility to structural similarity[J]. IEEE Transactions on Image Processing, 2004, 13(4): 600-612.

【5】Russo F, de Angelis A, Carbone P. A vector approach to quality assessment of color images[C]∥Instrumentation and Measurement Technology Conference Proceedings, 2008: 814-818.

【6】Chen J C. Application of ICA and BT-SVM in stereo image quality assessment system[D]. Tianjin: Tianjin University, 2012: 41-45.
程金翠. ICA和BT-SVM在立体图像质量评价系统中的应用[D]. 天津: 天津大学, 2012: 41-45.

【7】Wang G H, Li S M, Zhu D, et al. Application of extreme learning machine in objective stereo scopic image quality assessment[J]. Journal of Optoelectronics·Laser, 2014, 25(9): 1837-1842.
王光华, 李素梅, 朱丹, 等. 极端学习机在立体图像质量客观评价中的应用[J]. 光电子·激光, 2014, 25(9): 1837-1842.

【8】Bai J J, Sun Q, Jing S B, et al. Robust extreme learning machine and its application in analysis of near infrared spectroscopy data[J]. Laser & Optoelectronics Progress, 2015, 52(10): 103002.
白俊健, 孙群, 井诗博, 等. 稳健极限学习机及其在近红外光谱分析中的应用[J]. 激光与光电子学进展, 2015, 52(10): 103002.

【9】Goodfellow I, Bengio Y, Courville A. Deep learning[M]. Massachusetts, USA: The MIT Press, 2016: 331-339.

【10】Ciresan D, Meier U, Masci J, et al. Multi-column deep neural network for traffic sign classification[J]. Neural Networks, 2012, 32(1): 333-338.

【11】Lv Y, Yu M, Jiang G, et al. No-reference stereoscopic image quality assessment using binocular self-similarity and deep neural network[J]. Signal Processing Image Communication, 2016, 47: 346-357.

【12】Cheng L Y, Mi G Y, Li S, et al. Quality diagnosis of joints in laser brazing based on principal component analysis: support vector machine model[J]. Chinese Journal of Lasers, 2017, 44(3): 0302004.
程力勇, 米高阳, 黎硕, 等. 基于主成分分析-支持向量机模型的激光钎焊接头质量诊断[J]. 中国激光, 2017, 44(3): 0302004.

【13】Li S M, Lei G Q, Fan R. Depth maps super-resolution reconstruction based on convolutional neural networks[J]. Acta Optica Sinica, 2017, 37(12): 1210002.
李素梅, 雷国庆, 范如. 基于卷积神经网络的深度图超分辨率重建[J]. 光学学报, 2017, 37(12): 1210002.

【14】Lecun Y, Bottou L, Bengio Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.

【15】Maas A L, Hannun A Y, Ng A Y. Rectifier nonlinearities improve neural network acoustic models[C]∥Proceedings of 30th International Conference on Machine Learning, 2013, 30(1): 3.

【16】Krizhevsky A, Sutskever I, Hinton G. ImageNet classification with deep convolutional neural networks[C]. International Conference on Neural Information Processing Systems, 2012: 1097-1105.

【17】Srivastava N, Hinton G, Krizhevsky A, et al. Dropout: a simple way to prevent neural networks from overfitting[J]. Journal of Machine Learning Research, 2014, 15(1): 1929-1958.

引用该论文

Li Sumei,Chang Yongli,Duan Zhicheng. Objective Assessment of Stereoscopic Image Comfort Based on Convolutional Neural Network[J]. Acta Optica Sinica, 2018, 38(6): 0610003

李素梅. 基于卷积神经网络的立体图像舒适度客观评价[J]. 光学学报, 2018, 38(6): 0610003

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF